Skip to main content
Pre-release

This release is a pre-release and may not be stable for production use.

datasette-agent-modeler

PyPI Changelog Tests License

3D modelling tools for Datasette Agent - create and edit 3D models from chat

Screenshot

A Datasette Agent chat session - user has prompted "put the cat in the pot" - result is a 3D rendered cat sitting in a pot in a panel that allows for rotation of the model.

Installation

Install this plugin in the same environment as Datasette.

datasette install datasette-agent-modeler

Usage

This plugin adds 3D modelling tools to Datasette Agent. Ask the agent to build something - "make me a 3D model of a rocket ship" - and it will assemble one from primitive shapes and display it inline in the chat in an interactive viewer (drag to orbit, scroll to zoom, right-drag to pan). Then ask for changes - "make the fins purple and twice as tall" - and the agent edits the stored model and shows the updated result.

Four tools are registered with the agent:

  • create_model(name, objects, background?) - creates a model from a list of primitive shapes (box, sphere, cylinder, cone, torus, capsule, plane), each with optional position, rotation (degrees), scale, color and opacity. Returns a model_id.
  • edit_model(model_id, operations) - applies an ordered list of edit operations atomically: add_object, update_object (changed fields only - params are merged), remove_object, set_name, set_background.
  • get_model(model_id) - returns the current JSON document and revision number.
  • list_models() - lists stored models.

Models are stored in Datasette's internal database. Every edit saves a full snapshot as a new revision, and each viewer embedded in the chat transcript shows the model as it was at that revision, so the conversation doubles as a visual edit history.

The viewer is a <datasette-model-3d> custom element that renders with three.js (loaded from esm.sh), so browsing models requires internet access from the browser.

Permissions

The plugin registers a datasette-agent-modeler permission which gates all four tools - actors without it never see them. The --root user holds it by default:

datasette mydata.db --internal internal.db --root

See DESIGN.md for the full design notes, including the model document format and the 3D library research.

Development

To set up this plugin locally, first checkout the code. You can confirm it is available like this:

cd datasette-agent-modeler
# Confirm the plugin is visible
uv run datasette plugins

To run the tests:

uv run pytest

Metadata

Release files for datasette-agent-modeler 0.1a0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for datasette-agent-modeler 0.1a0
File Size Uploaded
datasette_agent_modeler-0.1a0.tar.gz 18.8 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for datasette-agent-modeler 0.1a0
File Interpreter ABI Platform
datasette_agent_modeler-0.1a0-py3-none-any.whl Python 3 none any Details

Total release size: 35.2 kB

Release files / datasette_agent_modeler-0.1a0.tar.gz

Download URL datasette_agent_modeler-0.1a0.tar.gz
Size 18.8 kB
Tags Source
SHA-256 checksum
How to use checksums
1bbd5e39c6735fcfe364f02c58cb3aac54e9c5fec3eeea57a61152bc3965f5de
BLAKE2b-256 checksum
How to use checksums
4761ba64566e5c77688fcd97a983476ceddc2715607ceb87a25cdfb1d78c6fbc
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Jul 29, 2026.

Transparency log

Release files / datasette_agent_modeler-0.1a0-py3-none-any.whl

Download URL datasette_agent_modeler-0.1a0-py3-none-any.whl
Size 16.4 kB
Tags Python 3
SHA-256 checksum
How to use checksums
49f6eb6afa3232c75dd2227702570fefc1bd03ac10a59246aef0ac20b42849f4
BLAKE2b-256 checksum
How to use checksums
c9f40c11bf40c3f08cab8b5d51c68dc03fc2b2ee6f25b111bd3702052dd4445a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Jul 29, 2026.

Transparency log

Release history Release notifications | RSS feed

This release

0.1a0 This release

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page